While most children with Wilms tumor (WT) are cured, a subset with aggressive disease continues to fail all current therapies. This proposal focuses on the role of vascular endothelial growth factor (VEGF) in WT angiogenesis. Our general strategy will be to characterize the response to altered status of VEGF in a xenograft model, focusing on changes in endothelium, recruited perivascular cells, and expression of angiogenesis-related genes (VEGF, angiopoietins, platelet-derived growth factor-B (PDGF-B), and Eph/Ephrins). Our preliminary results demonstrate that VEGF blockade initially inhibits WT xenografl growth and angiogenesis. However, prolonged treatment leads to recurrent tumor growth despite continued VEGF blockade, associated with profound alterations in vasculature. We hypothesize that a potential mechanism of this apparent resistance is the remodeling of tumor vessels to a state where they are less dependent on VEGF function. We will examine this mechanism by studying genes contributing to arterial/venous specification, vascular integrity, and remodeling (Aims 1 and 2) and the her2/neu oncogene, which may contribute to WT response to VEGF blockade (Aim 3). In Aim 1, we hypothesize that VEGF, in cooperation with ephrins, angiopoietins, and PDGFB, plays a critical role in forming vasculature with specific features that support aggressive behavior in WT. We will examine the response to VEGF blockade or overexpression in developing xenografts of different histology, relating tumor status to changes in vessel structure and expression of angiogenesis-related genes. We will determine if the regression of established vessel networks by VEGF blockade will alter specific vascular attributes. In Aim 2, we hypothesize that PDGF-B contributes to acquired resistance of WT xenografts to VEGF blockade. We will overexpress and block PDGF-B in Wilms tumor, and characterize the effect of altered status of PDGF-B on the response to VEGF antagonism. In Aim 3, we will determine whether her2/neu expression confers a relative survival advantage during VEGF blockade, by comparing the effects of blockade in xenografts expressing different levels of this receptor. We will characterize the effects of her2/neu activation and blockade on angiogenic genes in tumors in vitro and angiogenesis in vivo, and determine whether her2/neu inhibition affects acquisition of resistance to VEGF antagonists. The resulting preclinical data may provide a rational basis for use of anti-angiogenic therapies in Wilms tumor, and assist in circumventing their limitations.